The goal of this project is to identify and characterize clinically actionable mutations in all known breast cancer genes so that they can be incorporated into clinical care. DNA samples from 3000 cases and 2000 ancestry matched controls will be sequenced for all known breast cancer genes. Germline mutations will be evaluated with respect to function and relative risk of breast cancer. The discovery and characterization of clinically actionable variants in breast cancer genes involves three aims. In AIM 1, we will sequence all 18 known breast cancer genes in DNA samples from 3000 women with breast cancer and 2000 ancestry-matched controls. In parallel, we will similarly evaluate 6 new candidate genes for inherited breast cancer that have emerged from our independent family studies. For all characterizations, we will use our recently developed approach for simultaneous capture and multiplexed sequencing, to >300-fold median coverage, of exons, introns, regulatory, and flanking intergenic regions of breast cancer genes. In AIM 2, we will identify all variants predicted to lead to loss of gene function, including truncating mutations (whether nonsense, frameshifts, disrupting CNV and so on), complete genomic deletions, and putatively damaging missense mutations. We will assess all variants for effect on gene function using bioinformatics tools. Variants with possible splice effects will be evaluated by RT-PCR of patient RNA and ex vivo minigenes. The most promising missense alleles will be tested experimentally for DNA damage response, loss of enzyme activity and other functional assays as appropriate. In AIM 3, we will test the strength of association of breast cancer with each of the 24 genes, based on the combined frequency of unambiguous loss-of-function mutations in cases vs controls. Relative risks will be estimated for each gene. The challenge of integrating genomics into clinical care is timely now for inherited predisposition to breast cancer, because multiple causal genes are known, specialized breast screening modalities are available, risk- reducing surgery is effective, and treatment of breast cancer is influenced by the patient's genotype. The goal of this project is to provide the information and technology necessary for this translation.